A demonstration of K-Means clustering with and without dimensionality reduction using PCA, showcasing the impact of feature reduction on clustering results.

This module analyzes and clusters a dataset of cryptocurrencies based on price change percentages over different timeframes. Using K-means clustering, Principal Component Analysis (PCA), and the elbow method, this project aims to find optimal clusters for understanding cryptocurrency behavior.

컴퓨팅사고와 데이터분석 기초 최종 프로젝트 (2024-1)

Comprehensive examination of best-selling books, focusing on understanding sales patterns, genre distributions, and the impact of various features on book performance.This project aims to predict book sales and classify genres, providing valuable insights for authors, publishers, and readers.

Analyzed global population data using decision tree regression, calculating growth rates and evaluating model performance with RMSLE for insights.

Analyzes driver retention on the Ola platform. Identifies churn-prone drivers and offers insights to improve retention strategy, reducing costs related to recruitment and onboarding

A capstone project in collaboration with Zama to develop a privacy-preserving machine learning model using PPML, FHE and Concrete ML to detect banking frauds.

This project profiles high-risk "White Collar" and "Grey Collar" customers to assess their creditworthiness. It aims to balance profitability and customer satisfaction using data science techniques for better loan approval decisions.

A repository housing a CNN model for text recognition, implemented in Python with TensorFlow and OpenCV.

Vidéosurveillance intelligente en environnement confus

Uncovered keyword frequency and correlation with engagement metrics using NLTK library to identify impactful keywords and phrases.

This repository provides an implementation for the data pipelines and AI models used in the COPERIA project.

Depression Prediction and Analysis

Meu progresso no livro Análise Prática de Séries Temporais: Predição com Estatística e Machine Learning

LSTM for stock and crypto prices predictions

This Django-based movie recommendation system utilizes TF-IDF and cosine similarity to provide personalized movie suggestions based on user ratings and genre preferences. It enhances the user experience by helping them discover films that match their interests. The project also uses SQLite for data storage and includes user authentication.

Repo of the "Hands-on Machine Learning with Scikit-Learn, Keras and TensorFlow (3rd edition)" book.

This classifier predicts the genre of books based on titles or descriptions using a Machine Learning model trained on an Amazon books dataset.

Repo for storing notes and pracitcal implemtations

Learning Machine Learning By using TensorFlow From Scratch